You are so right. I intend to write something short along that lines for the next instance of the CP course I am teaching.
I will switch from Gecode/J to Gecode and hence it will be more rewarding to explain Gecode. Christian -- Christian Schulte, www.ict.kth.se/~cschulte/ -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Denys Duchier Sent: Friday, December 14, 2007 12:08 AM To: [EMAIL PROTECTED] Subject: Re: [gecode-users] inter_r with views of vars in different spaces "Christian Schulte" <[EMAIL PROTECTED]> writes: > On a more general level, performing "read" operations (anything that is > const) on a view or variable is independent of the space the view or > variable belongs to (the only thing not to be done is to delete the space). > Anything goes! > > When performing a "write" operation (anything that is not const) one has to > use the space as argument to these operations to which the view or variable > belongs. That's all. Thank you, Christian, this is the sort of high level design principle that I miss most in the current documentation. My major problem with this existing documentation is that it does not offer a "big picture"; and, lacking that, gecode is a lot like "adventure" (and all you young people, keep quiet ;-), "a maze of twisty little passages all alike", but without the fun ('cause I don't have the time anymore, unfortunately). Is it foolish to imagine that a L3 or M1 student (or whatever the corresponding levels are in Sweden) could start a wiki documenting (1) design principles, and (2) case uses? I would be so greatful. ... if this is stupid... just ignore me Cheers, --Denys _______________________________________________ Gecode users mailing list [EMAIL PROTECTED] https://www.gecode.org/mailman/listinfo/gecode-users _______________________________________________ Gecode users mailing list [EMAIL PROTECTED] https://www.gecode.org/mailman/listinfo/gecode-users
